语音识别系统

实验结果表明所提算法可以有效提高语音识别系统的识别正确率。研究了基于多次自相关运算的去噪算法，其目的是保证在去噪的同时而不改变语音信号的频谱结构。

Experimental results reveal that the proposed speech enhancement algorithm can improve recognition ratio on ASR system . A denoising algorithm based on multi-order autocorrelation was studied , which is used to retain the structure of speech frequency spectrum while suppressing the noise .

连续语音识别系统中的OnePass搜索方法

One Pass Search Algorithm in Contnuous Speech Recognition System

基于SYSTEMC系统级的语音识别系统性能分析库函数的设计

Design of the Library Function of Performance Analysis on Speech Recognition Based on System C

基于MATLAB和BP网络的语音识别系统

The speech recognition system based on MATLAB and BP neural networks

基于DSP的非特定人语音识别系统

Research and Realization of Speaker-Independent Speech Recognition System based on DSP

小词汇量孤立词语音识别系统的DSP实现

DSP Implementation of Speech Recognition System for Small Vocabulary Isolated Words

连接数字语音识别系统的DSP实时实现

DSP Real Time Implementation of Speech Recognition Based on Connected Digits

基于DSP的语音识别系统及其结构

A Speech Recognition System and Structure Based on the DSP

小词表实时语音识别系统的定点DSP实现

Fixed Point DSP Application of A Small Glossary Real Time Speech Recognition System

基于DSP技术的汉语数码语音识别系统

Mandarin Digit Speech Recognition System Based on DSP Technology

MATLAB环境下的语音识别系统

Speech Recognition in MATLAB Environment

文中论述了动态时间归整理论和隐马尔可夫模型原理，用MATLAB语言编程研究了它们在语音识别系统中的应用。

The DTW theory and HMM theory are discussed . Their applications in recognition are analyzed through the MATLAB programs .

本课题首先建立了一个基于隐马尔科夫模型（HMM）的孤立词语音识别系统。

An isolated-word speech recognition system based on HMM has been established .

基于连续HMM语音识别系统的构建与分析

Construction and Analysis of Speech Recognition System of Continual HMM

HMM非特定人孤立词语音识别系统的片上实现

On Chip Realization of HMM Speaker-independent Isolated Word Speech Recognizer

文章构造了基于经典HMM模型的汉语连续语音识别系统。

A baseline system based on classical HMM is implemented in this paper .

动态时间规整算法DTW（DynamicTimewarping）作为一种非线性时间匹配技术已成功地应用于语音识别系统中。

Dynamic Time Warping ( DTW ) has been widely used in speech recognition sys-tems as a , nonlinear time alignment technique .

第五章阐述了本文所建立的基于HMM建模的蒙古语连续语音识别系统的框架及其实现以及蒙语语音识别应用技术研究。

In Chapter five , I introduce a Mongolian speech recognition system basing on HMM.

HMM在语音识别系统中的应用

Application of HMM in Automatic Speech Recognition System

SOM结合MLP的神经网络语音识别系统

A New Speech Recognition System Using SOM and MLP Neural Networks

主要内容是建立一种基于RBF神经网络的语音识别系统，探讨RBF神经网络在语音识别中的应用。

It presents a speech recognition method , which is based on Radial Basis Function Neural Networks ( RBFNN ) .

本文中笔者设计的语音识别系统选择STRAIGHT谱作为语音特征参数，用差别子空间法进行模板训练。

In the system designed in this paper , STRAIGHT spectrum is selected as the identifiable feature , and the difference subspace is for template training .

最后给出了一个以DSP为核心的连续数字语音识别系统，以说明DSP在该系统中发挥的核心作用。

What , more , the thesis offers a whole system of . continuous digital speech recognition to show the core function of DSP in it .

本文建立了一个孤立词语音识别系统，并利用MATLAB的语音工具箱voicebox对系统进行了仿真和分析。

This article has established an isolated-word speech recognition system , and has carried on the simulation and analysis using MATLAB pronunciation toolbox voice box to the system .

基于DCOM的分布式电话语音识别系统

Distributed Telephony Speech Recognition System Based on DCOM

应用概率LR文法分析器于语音识别系统

Speech recognition based on probabilistic LR parser

然而实验室中有效的语音识别系统在鲁棒性（Robustness）、灵活性、自适应性方面远不能满足实际应用的需要，特别是当前语音识别技术中三大难点之一的噪音问题是阻碍实际应用的关键。

But the performance of the speech recognition system degraded greatly under actual environment because of weakness in robustness , flexibility and self-adaptive , especially the noisy problem .

为了验证Viterbibeam搜索算法的有效性，本文设计了可行的Viterbibeam搜索策略，构建了小型英文连续语音识别系统ATW（AskTheWay）。

To verify Viterbi beam search algorithm and pruning strategies , we set up a small English CSR system called Ask The Way ( ATW ) .

提出了基于肌电信号（EMG）的无声语音识别系统。

It is proposed in this paper that electromyography ( EMG ) signal can be used in unvoiced speech recognition systems .

参数共享是基于隐Markov模型（HiddenMarkovmodel，HMM）的语音识别系统的参数训练中的一个关键性问题，因此在语音识别的诸多领域中都有重要的应用。

Parameter sharing is a key element in the training of speech recognition systems based on hidden Markov model ( HMM ) with important applications in many fields of speech recognition .